False-Positive Pregnancy Result in a Patient with Bone Mass Haoliang Xu, 1 Mohamed Rizwan Haroon Al Rasheed, 1 Elizabeth L. Wiley, 1 and Ming Jin 1 * CASE DESCRIPTION An 18-year-old woman presented to our institution complaining of a painful left shoulder. She described it as a sharp and throbbing pain that had begun insidiously but had progressively worsened to involve most of her left shoulder. She also noted that the pain would occasionally radiate down her arm and that she had difficulty mobilizing her shoulder, resulting in a pseudo-paralysis of the left upper extremity. She denied associated symptoms such as weight loss, anorexia, and fever. The remainder of her history was noncontributory. Except for the limited range of motion, her physical examination, including motor strength and sensation, was unremarkable. Radiologic imaging of her left shoulder demonstrated a large lytic bone lesion located on the proximal head of the left humerus with extension through the greater tuberosity to involve the deltoid muscle. A bone core needle biopsy demonstrated a mixture of mononucleated and multinucleated giant cells compatible with chondroblastoma, giant cell tumor, and Langerhans cell histiocytosis. As the mass had completely replaced the proximal head of the humerus, the patient underwent resection of the mass. During preoperative evaluation, her serum calcium was 8.4 mg/dl (reference interval, 8.6 10.6) and urine human chorionic gonadotropin (hcg) 2 was positive (Combo Rapid Test B1077-23, Cardinal Health) with the confirmatory serum hcg of 38 IU/L [reference interval for nonpregnant female: <5, Beckman UniCel DxI 800, standardized to WHO 5th International Standard (IS)]. Because the patient denied sexual activity, a potentially false-positive result was investigated. The increased hcg was confirmed by our chemistry laboratory (Laboratory A) and 2 other local clinical laboratories (Laboratory B and Laboratory C) using immunoassays of Abbott Architect i2000 (standardized to WHO 4th IS) and Roche Cobas 602 [standardized to National Institute for Biological Standards and Control (NIBSC) 4th IS]. However, wide interassay variation (ranging from 19 to 40 IU/L) was noticed among these 3 immunoassays. Meanwhile, a serum sample from a known healthy female in early pregnancy was tested concurrently in these 3 laboratories as a positive control, all of which resulted in comparable hcg concentrations of 48 IU/L. Twenty-four hours after tumor resection, the patient's serum hcg concentration dropped to 2 IU/L and remained below this concentration throughout her hospital stay. All 3 laboratories detected identical low serum hcg concentrations after tumor resection (Table 1). These results suggest that the patient was not pregnant, and the mass lesion produced specific isoforms of paraneoplastic hcg, which were recognized differently by 1 Department of Pathology, University of Illinois at Chicago, Chicago, IL. *Address correspondence to this author at: Department of Pathology (MC 847), University of Illinois at Chicago, 840 South Wood St., Chicago, IL 60612. Fax 312-996-7586; e-mail mjin@uic.edu. DOI: 10.1373/jalm.2017.022988 2017 American Association for Clinical Chemistry 2 Nonstandard abbreviations: hcg, human chorionic gonadotropin; IS, International Standard; GCTB, giant cell tumor of bone; TGFβ, transforming growth factor β; hcgn, nicked hcg; hcgβcf, core fragment of β-hcg; hcgβn, nicked β-hcg subunit. September 2017 02:02 000 JALM 1 Copyright 2017 by American Association for Clinical Chemistry.
Table 1. Comparison of the serum hcg concentration of pre- and postsurgical excision of GCTB with 3 different immunoassays. hcg of patient, IU/L (days post-surgery) hcg of healthy pregnant female, IU/L Immunoassays Isoform recognition a 0 day 1 day 3 days Beckman hcg, hcgβ, hcgn, hcgβn 40 2 1 48 Abbott hcg, hcgβ 19 2 Not available 48 Roche hcg, hcgβ, hcgn, hcgβcf 32 2 Not available 48 a hcg isoforms: intact hcg, hcg; β-hcg subunit, hcgβ; nicked hcg, hcgn; nicked hcgβ, hcgβn; core fragment of β-hcg, hcgβcf. the 3 immunoassays because of their variable recognition spectra. Histopathologic evaluation of the resected sample revealed 2 cell populations composed of mononuclear stromal cells and osteoclast-type giant cells with strong positivity for osteonectin immunohistochemical stain in the mononuclear cells but not the osteoclast-like giant cells. Therefore, a final diagnosis of giant cell tumor of bone (GCTB) was made (Fig. 1, A C). Moreover, strong hcg immunohistochemical expression was noted in about 1% of the mononuclear cells and rarely in smaller multinucleated giant cells (Fig. 1D). DISCUSSION GCTB is mostly a benign, locally aggressive neoplasm that rarely metastasizes. It occurs more commonly in women with a peak incidence in the third and fourth decade, representing 4% 5% of primary neoplasms of bone (1). It is often located at the end of long bones and presents clinically with pain, swelling, limitation of motion, and/or fracture. Grossly, it appears hemorrhagic and soft with secondary cystic areas. Histologically, it is composed of neoplastic mononuclear cells interspersed among a large population of nonneoplastic osteoclast-type multinucleated giant cells (1, 2). hcg is a glycoprotein that is predominately secreted by the trophoblastic cells of the placenta after implantation of the fertilized ovum and is primarily used for pregnancy testing. Enhanced β-hcg expression has been linked to advanced stage and grade in certain diseases, such as urothelial carcinomas (3). Although the underlying mechanisms are still unknown, current proposed theories point to the autocrine/paracrine antiapoptotic character of β-hcg in neoplastic cell development. β-hcg is structurally homologous to transforming growth factor-β (TGFβ) and may bind to a component of the TGFβ receptor complex blocking the TGFβ-initiated apoptosis signaling pathway (4). Thus, in clinical practice, β-hcg may serve not only as a diagnostic indicator, but also as an important prognostic indicator for neoplastic diseases. Lawless et al. (2) first reported that a significant number of GCTB expressed β-hcg by immunohistochemical staining and noted that some GCTB may also produce increased serum and urine β-hcg. In fact, immunohistochemical expression of β-hcg was stronger and more diffuse in proliferative and infiltrative areas of the tumor and correlated with a trend of tumor recurrence and metastasis. Interestingly, although the majority of GCTB patients in the study by Lawless et al. showed β-hcg expression in tumor cells, only 2 patients had increased β-hcg in urine and/or serum (2). This discordance is not surprising because the sporadically distributed tumor cells may secrete hcg at a rate that precludes hormone accumulation to a detectable concentration. Nevertheless, in the study by Fitzhugh et al. (5) and our current report, the concurrent elevation of urine and serum hcg, as well as tissue hcg expression, 2 JALM 000 02:02 September 2017
CASE REPORT Fig. 1. Giant cell tumor of bone. (A and B), Histology (hematoxylin and eosin) of the resected lesion demonstrates mononuclear cells (arrow) interspersed between giant cells (arrowhead; A: 100; B: 400). (C), Osteonectin immunohistochemistry shows diffuse positive staining in the mononuclear cells ( 200). (D), hcg immunohistochemistry demonstrates focal positivity in the mononuclear cells (arrow) and smaller giant cells (arrowhead; 200). can be detected in GCTB patients. After tumor excision, serum β-hcg concentrations returned to baseline, confirming the tumor origin of hcg. Immunoassays are widely available for the quantitative measurement of hcg in serum. Although these assays are sensitive and accurate in clinic practice for pregnancy early detection and serial follow-up, interpretation of the testing results may be challenging under certain clinic scenarios. For example, false-positive results can occur because of heterophile antibody and medications (6). Moreover, serum hcg elevation has been found in a number of gestational trophoblastic diseases, such as hydatiform mole, choriocarcinoma, and nontrophoblastic neoplasms (7). In addition, elevation of β-hcg is one of the strongest independent September 2017 02:02 000 JALM 3
prognostic indices in patients with pancreatic cancer (8). Ectopic β-hcg expression in bone tumors is associated with worse prognosis. β-hcg secreting tumor cells have been demonstrated in approximately 50% 60% of osteosarcomas, with occasional β-hcg elevation in urine that decreased to an undetectable concentration after chemotherapy and definitive resection (9). Although β-hcg expression in GCTB has been reported (2, 5), because of its low incidence, hcg expression, alteration of serum and/or urine concentrations, and the composition of its isoforms in GCTB patients are not well documented. In this case study, we examined the hcg serum concentrations by using 3 immunoassays with variable hcg isoform recognition spectra. Each individual immunoassay has specific antibody-hcg binding sites that are specific and sensitive for certain isoforms of hcg. These 3 different assays yielded identical values when they were applied to a healthy pregnant patient, indicating their compatibility when testing for intact hcg and β-hcg (2 major hcg isoforms during pregnancy). This result is consistent with the data of CAP (College of American Pathologists) Survey Ligand Assay General K-A 2016, in which the 3 assays show similar performance at low concentrations, while variations are noted at high concentrations. On the contrary, a wide interassay variation emerged when testing the patient's presurgical serum. The lowest value was reported by the Abbott immunoassay (19 IU/L), which is structured to detect hcg and β-hcg subunit. Meanwhile, both Beckman and Roche immunoassays detected higher hcg values (40 and 32 IU/L, respectively) and showed wider recognition spectra toward hcg isoforms. Specifically, Beckman immunoassay detected the additional isoforms of nicked hcg (hcgn) and nicked β-hcg subunit (hcgβn), while Roche immunoassay detected hcgn and core fragment of β-hcg (hcgβcf). The discrepancy of the hcg values of the patient by the 3 immunoassays may be due to the other tumor-derived hcg isoforms, or TAKEAWAYS GCTB, a tumor that often occurs insidiously in younger females, may produce hcg, leading to a false-positive pregnancy test. It is important to be aware of hcg expression by this tumor to avoid misdiagnosis of pregnancy during preoperative workup. The predominant isoforms of ectopic hcg by GCTB causes an interassay variation with different hcg immunoassay instruments. Therefore, a consistent or standardized hcg immunoassay targeting specific hcg isoforms is required for following up patients with GCTB. hcg may serve as an ancillary marker of GCTB in diagnosis, posttreatment followup, and monitoring for recurrence. posttranslational modification. Further studies to explore GCTB-derived hcg isoforms including hcgn and hcgβn, as well as posttranslational modification including hyperglucosylation, sulfation, and degradation of hcg, are needed to understand the nature of hcg by GCTB. hcgβcf is an unlikely candidate for causing this variation because it is detectable in the urine only after being exclusively degraded and excreted in the kidney. Indeed, testing these isoforms may have important diagnostic and prognostic value in patients with certain neoplastic diseases and could be particularly important in clinical practice depending on the prevalent form or proportion of hcg in the diseases (7). For instance, in lung cancer, the use of urine hcgβcf as a biochemical marker showed sensitivity of 48% in the early stages and 72% in most advanced cases (10). Until now, the alteration of hcg expression and predominate isoforms in GCTB patients has not been investigated, except for the 2 reports from Lawless et al. (2) and Fitzhugh et al. (5). However, in both these studies, 4 JALM 000 02:02 September 2017
CASE REPORT β-hcg was arbitrarily designated and evaluated as the only hcg isoform produced ectopically by the tumor cells. Our case indicates that there may be other hcg isoforms, besides β-hcg, that are produced by tumor cells in patients with GCTB. This is the first report of such a finding in the literature. In conclusion, we report a unique case of GCTB, with concomitant increased serum hcg, exhibits wide interassay variation. Because GCTB often occurs in younger females with an insidious onset, it is important to be aware that hcg expression by GCTB may possibly cause a false-positive pregnancy result. Although proper validation would be necessary, our results also suggest that hcg may serve as an ancillary marker of GCTB in diagnosis and postsurgical resection follow-up. Author Contributions: All authors confirmed they have contributed to the intellectual content of this paper and have met the following 4 requirements: (a) significant contributions to the conception and design, acquisition of data, or analysis and interpretation of data; (b) drafting or revising the article for intellectual content; (c) final approval of the published article; and (d) agreement to be accountable for all aspects of the article thus ensuring that questions related to the accuracy or integrity of any part of the article are appropriately investigated and resolved. Authors Disclosures or Potential Conflicts of Interest: No authors declared any potential conflicts of interest. REFERENCES 1. Athanasou N, Bansal M, Forsyth R, Reid R, Sapi Z. Giant cell tumor of bone. In: Fletcher CD, Bridge J, Hogendoorn P, Mertens F, editors. WHO classification of tumours of soft tissue and bone. 4th ed. Lyon: IARC Press; 2013. p 321 4. 2. Lawless ME, Jour G, Hoch BL, Rendi MH. Beta-human chorionic gonadotropin expression in recurrent and metastatic giant cell tumors of bone: a potential mimicker of germ cell tumor. Int J Surg Pathol 2014;22:617 22. 3. Dirnhofer S, Koessler P, Ensinger C, Feichtinger H, Madersbacher S, Berger P. Production of trophoblastic hormones by transitional cell carcinoma of the bladder: association to tumor stage and grade. Hum Pathol 1998; 29:377 82. 4. Iles RK. Ectopic hcgbeta expression by epithelial cancer: malignant behaviour, metastasis and inhibition of tumor cell apoptosis. Mol Cell Endocrinol 2007;260 2:264 70. 5. Fitzhugh VA, Katava G, Wenokor C, Roche N, Beebe KS. Giant cell tumor of bone with secondary aneurysmal bone cyst-like change producing β-human chorionic gonadotropin. Skeletal Radiol 2014;43:831 4. 6. Stenman UH, Tiitinen A, Alfthan H, Valmu L. The classification, functions and clinical use of different isoforms of HCG. Hum Reprod Update 2006;12:769 84. 7. de Medeiros SF, Norman RJ. Human choriogonadotrophin protein core and sugar branches heterogeneity: basic and clinical insights. Hum Reprod Update 2009;15:69 95. 8. Louhimo J, Alfthan H, Stenman UH, Haglund C. Serum HCG beta and CA 72 4 are stronger prognostic factors than CEA. CA 19 9 and CA 242 in pancreatic cancer. Oncology 2004;66:126 31. 9. Lee AF, Pawel BR, Sullivan LM. Significant immunohistochemical expression of human chorionic gonadotropin in high-grade osteosarcoma is rare, but may be associated with clinically elevated serum levels. Pediatr Dev Pathol 2014;17:278 85. 10. Yoshimura M, Nishimura R, Murotani A, Miyamoto Y, Nakagawa T, Hasegawa K, et al. Assessment of urinary beta-core fragment of human chorionic gonadotropin as a new tumor marker of lung cancer. Cancer 1994;73: 2745 52. September 2017 02:02 000 JALM 5